|
|
most, if not all of these responses seem to be more geared towards a ring game format rather than a tournament. (posts posted while i was writing this have emphasized the difference between tourney and ring) I think that a good way to reduce overall variance in tournaments might be to not let people draw for even remotely the right price (barring them drawing dead). you may be leaving some chips on the table if you overbet 2 pair on a draw heavy flop, but all of those chips in the middle are now in your stack for this tournament. you can now (hopefully) leverage these extra chips in later situations to further increase your chipstack, thereby making up for the chips left "on the table" in an earlier hand.
I am not advocating this type of play, nor am i saying it's the correct way (or even the way i play) to play a SNG, but it would seem to remove a lot of the 'luck' factor from your SNG variance. if you only go to the river with chips left behind and a <5% chance that the opp can win the hand, then you should vastly reduce the number of times you bust out of a SNG due to being outdrawn. rather than just not giving odds for a flush draw, and making an few extra chips 2/3 of the time, take all of the chips in the middle 90% of the time, because you will still get some people who cant let go of a draw. it would be closer to 80-85% for most of the 5.50 tournaments ive played in.
eg. starting stacks of 1500, 9 players. say you face an average of 12 flush draws against per tournament. I have no idea what the actual number would be, this was chosen at random. they would be spread out over the first four blind levels, weighted towards the first two. we will also assume starting pot of 3.5xBB (this includes your call of the blind). also assumes only 1 opp per hand, and that flush hits on the river, not the turn.
10/20: 5 flush draws faced.
starting pot (before flop) = 70. for this pot, a bet of 40 on the flop, plus a bet of 80 on the turn is sufficient to give a flush draw bad odds. final pot at the river = 310. amount you've invested = 140. net amount per flush draw = 112 chips (170*66). times this by 5 flush draws and we have increased our stack by 560 chips.
if you were to massively overbet this pot, lets say 210 into a 70 pot, this gives horrible odds to a flush draw, and has a good probability of taking the pot now. assume 80% folds. 4/5 times you will take 70 chips, = 280.
called: final pot at river = (assuming AI on turn) 3050. amount invested = 1500. net amount per flush draw seen until the river = 1023.
so, over all 5 flush draws faced at this level, we net 1303 by 3xpot on flop, AI on river, vs. 560 chips by simply giving someone bad odds. of course, there is a .06 chance that you will be eliminated after going AI on the turn, while no chance of being eliminated by playing standard. but, you have only faced 1 flush draw that got to see all 5 cards vs. 5 flush draws while playing standard.
this makes a lot of assumptions, and the math may be a little dodgy, but I think it makes my point. not letting draws in for anywhere near the right price should decrease in-tournament variance, which should ultimately lead to a decrease in SNG variance.
we have all lost to 3 flush draws in a row (or is it just me) and busted out of a tournament even though all 3 were played properly (ie. given bad odds to chase). if you instead give horrible odds and not worry about not getting the 'maximum' amount of chips off of every hand where you are facing a draw that has a decent probability of hitting, then you should stay in more tournaments. this should allow you, since you are presumably better than most of the competition that you are playing against, to money/win more tournaments simply because you are in them longer, and are playing more hands.
|